Medical vending platform mobile management method, device, equipment and storage medium

Abstract

The present invention relates to the technical field of medical vending platforms, and discloses a mobile management method, device, equipment and storage medium for medical vending platforms, wherein the method comprises: creating an initial product shelving plan based on vending machine operation data of the medical vending platform, and collecting medical product image data; performing residual learning and two-level feature extraction to obtain medical product feature data; inputting a medical product recognition model to perform medical product recognition and obtain medical product recognition results; performing inventory detection to generate product inventory data, and sending it to a mobile management terminal to generate a replenishment task instruction; inputting RFID tag information and the replenishment task instruction into a replenishment execution unit to generate replenishment operation data and update real-time inventory information; adjusting the initial product shelving plan based on real-time inventory information and historical sales data to generate a target product shelving plan, wherein the method improves the operation efficiency and service quality of the medical vending platform.

Description

Technical Field

[0001] The present invention relates to the technical field of medical vending platforms, and in particular to a mobile management method, device, equipment and storage medium for a medical vending platform. Background Art

[0002] Self-service medical supply vending services have become an important way to meet the public's demand for convenient medical supply purchases. Traditional medical vending platforms rely primarily on manual management, resulting in inefficiencies and poor real-time performance in areas such as product replenishment, inventory monitoring, and sales analysis, making it difficult to meet users' needs for timely purchases of medical supplies.

[0003] Existing mobile management systems for medical vending platforms generally rely on single RFID or barcode recognition technology for product management. This approach is prone to recognition errors when faced with a wide variety of products and complex placements. It also fails to effectively address anomalies such as misplaced or obstructed product placement, leading to inaccurate inventory information and compromising the effectiveness of replenishment decisions. Furthermore, current medical vending platforms lack intelligent mechanisms for optimizing product placement, making it impossible to dynamically adjust product placement strategies based on sales trends across different time periods and regions. Furthermore, they are unable to intelligently allocate replenishment tasks and optimize routes, resulting in wasted human resources and reduced platform operational efficiency. Summary of the Invention

[0004] The present invention provides a mobile management method, device, equipment and storage medium for a medical vending platform, which improves the operational efficiency and service quality of the medical vending platform.

[0005] In a first aspect, the present invention provides a mobile management method for a medical vending platform, the mobile management method for a medical vending platform comprising:

[0006] Create an initial product listing plan based on the vending machine operation data of the medical vending platform and collect medical product image data;

[0007] performing residual learning and two-level feature extraction on the medical product image data to obtain medical product feature data;

[0008] Inputting the medical commodity feature data into a medical commodity recognition model to perform medical commodity recognition and obtain a medical commodity recognition result;

[0009] Performing inventory detection based on the medical product identification result to generate product inventory data, and sending the product inventory data to the mobile management terminal to generate a replenishment task instruction;

[0010] Inputting the RFID tag information and the replenishment task instruction into the replenishment execution unit, generating replenishment operation data, and updating the real-time inventory information;

[0011] The initial product listing plan is adjusted based on the real-time inventory information and historical sales data to generate a target product listing plan.

[0012] In a second aspect, the present invention provides a mobile management device for a medical vending platform, the mobile management device for a medical vending platform comprising:

[0013] The acquisition module is used to create an initial product listing plan based on the vending machine operation data of the medical vending platform and collect medical product image data;

[0014] an extraction module, configured to perform residual learning and two-level feature extraction on the medical product image data to obtain medical product feature data;

[0015] an identification module, configured to input the medical commodity feature data into a medical commodity identification model to perform medical commodity identification and obtain a medical commodity identification result;

[0016] a detection module, configured to perform inventory detection based on the medical product identification result, generate product inventory data, and send the product inventory data to the mobile management terminal to generate a replenishment task instruction;

[0017] A generation module, configured to input the RFID tag information and the replenishment task instruction into a replenishment execution unit, generate replenishment operation data, and update real-time inventory information;

[0018] An adjustment module is used to adjust the initial product listing plan based on the real-time inventory information and historical sales data to generate a target product listing plan.

[0019] A third aspect of the present invention provides a mobile management device for a medical vending platform, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the mobile management device for the medical vending platform executes the above-mentioned mobile management method for the medical vending platform.

[0020] A fourth aspect of the present invention provides a computer-readable storage medium having instructions stored therein, which, when executed on a computer, enables the computer to execute the above-mentioned medical vending platform mobile management method.

[0021] In the technical solution provided by the present invention, accurate identification of medical products and real-time inventory monitoring are achieved through deep learning and computer vision technology, in conjunction with the RFID identification system; a two-stage feature extractor and residual learning method are adopted to effectively improve the accuracy of product identification in complex environments; by establishing a dynamic product shelving solution optimization mechanism, intelligent adjustment of product layout and automatic optimization of replenishment strategy are achieved; by introducing mobile management terminals and replenishment execution units, a complete replenishment-acceptance-shelfing closed-loop management system is constructed, which significantly improves the operational efficiency and service quality of the medical vending platform; through intelligent analysis of real-time inventory information and sales data, an adaptive optimization mechanism of the sales model is established, which realizes the rational allocation of product resources and effective control of operating costs.

[0022] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or understood by practicing the present invention. The purposes and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0023] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A schematic diagram of an embodiment of a mobile management method for a medical vending platform according to an embodiment of the present invention;

[0025] Figure 2 This is a schematic diagram of an embodiment of a mobile management device for a medical vending platform according to an embodiment of the present invention;

[0026] Figure 3 Schematic diagram of an embodiment of a mobile management device for a medical vending platform in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0028] The terms "including," "having," and any variations thereof, as used in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units not listed, or may optionally include other steps or units inherent to the process, method, product, or device.

[0029] To facilitate understanding of this embodiment, a mobile management method for a medical vending platform disclosed in an embodiment of the present invention is first described in detail. Figure 1 As shown, this method includes the following steps:

[0030] 101. Create an initial product listing plan based on the vending machine operation data of the medical vending platform and collect medical product image data;

[0031] It is understandable that the execution subject of the present invention can be a medical vending platform mobile management device, or a terminal or a server, which is not limited here. The embodiment of the present invention is described by taking the server as the execution subject as an example.

[0032] Specifically, data cleaning is performed on the sales records of each vending machine in the medical vending platform to remove invalid data, erroneous data, and outliers, ensuring that the resulting sales records reflect actual sales performance. After data cleaning, valid sales data is obtained and classified by product category and sales time to generate categorized sales data. Sales performance is then broken down into specific product categories and sales time periods. Time series analysis is performed on the categorized sales data to determine sales trends and fluctuations within each time period, generating sales curves for each time period. By calculating the sales curves, the sales frequency of each time period is determined, reflecting the level of sales activity within that specific time period and deriving sales time period parameters. These sales time period parameters are input into a product turnover rate calculation model to calculate the turnover rate of each product. Product turnover rate reflects the product's sales cycle, i.e., the speed at which a product is sold from being put on the shelf. Based on this product turnover rate data, safety stock levels for each product are set to ensure sufficient inventory of each product in the vending machine to cope with demand fluctuations, thereby determining replenishment threshold parameters. Spatial distribution analysis is performed on the product placement data of each vending machine to generate popularity data for each product location. Product location popularity data primarily reflects the sales frequency of products at a specific location and customer interaction with that location. Based on this location popularity data, product placement is prioritized to generate location priority data. Products with high sales potential are placed in more visible and accessible locations. Correlation analysis is performed between location priority data and sales time period parameters to generate location-time period matching data. This location-time period matching data reveals which products, when placed, maximize sales within a specific time period. Based on this matching data, a product location distribution matrix is ​​generated, product layout parameters are derived, and the optimal placement of each product within the vending machine is determined, helping to improve product exposure and sales efficiency. An initial product placement plan is generated based on replenishment threshold parameters, product layout parameters, and sales time period parameters to maximize the sales potential of products within the vending machine and meet customer needs during different time periods. Furthermore, based on this initial product placement plan, the remaining quantity of each product within the vending machine is monitored in real time to identify any potential out-of-stock risks. By monitoring the remaining quantity of each product, a stock-out risk value is calculated for each product, which is then used to prioritize image acquisition. For products at higher risk of out-of-stock, image acquisition is prioritized to ensure timely identification of inventory changes and demand fluctuations. Based on product layout parameters, the product display within the vending machine is spatially mapped to obtain specific coordinate data for each product. Using this coordinate data, the image acquisition device's acquisition angle is set to ensure both the full view and details of each product are captured. The image acquisition device is controlled based on the image acquisition priority, performing adaptive angle acquisition to ensure the integrity and accuracy of the image data. This adaptive acquisition method is used to obtain image data for medical products.

[0033] 102. Perform residual learning and two-level feature extraction on the medical product image data to obtain medical product feature data;

[0034] Specifically, brightness analysis is performed on medical product image data to generate an image brightness distribution matrix, which describes the brightness distribution across various image regions. Based on this image brightness distribution matrix, the image data is segmented into multiple blocks, forming an image block sequence. During the residual learning process, targeted feature extraction is performed on different image portions, improving the model's adaptability and accuracy for different brightness regions. The image block sequence is then input into the backbone feature extraction layer of the residual network, which consists of three residual units. Each residual unit contains two 3×3 convolutional layers and a short-circuit connection. This short-circuit connection effectively alleviates the vanishing gradient problem in deep networks and enhances network training performance. A ReLU activation function is used in each convolutional layer to enhance the network's nonlinear representation capabilities, while batch normalization improves training stability and convergence speed. After processing by these three residual units, a backbone feature map is generated, which contains the basic feature information of the medical product. Residual mapping is then performed on the backbone feature map, further adjusting the features to capture more subtle feature variations in the image, resulting in a residual feature map. The residual feature map is element-wise added to the original sequence of image patches to produce enhanced image data. This enhanced image data is input to the first-level feature extraction network of the two-stage feature extractor. This feature extraction network adopts a pyramidal structure consisting of five convolutional layers. Each convolutional layer is followed by a max pooling operation. This design extracts product features at different scales, resulting in a multi-scale feature map. The multi-scale feature map contains product feature information at different spatial resolutions. Channel attention is performed on the multi-scale feature map to produce a channel weight matrix. By assigning different weights to different channels, important feature channels are highlighted and irrelevant or noisy channels are suppressed. The multi-scale feature map is weighted according to the channel weight matrix to produce the first-level feature data. This first-level feature data is input to the second-level feature extraction network of the two-stage feature extractor. The second-level feature extraction network uses four parallel dilated convolution branches with dilation rates set to 1, 2, 4, and 8, respectively. The use of dilated convolutions enables the network to increase the receptive field while using fewer parameters, capturing feature information over a wider range, and producing multi-receptive field feature maps. In order to improve the effectiveness of the multi-receptive field feature map, a spatial attention calculation is performed on it to obtain a spatial weight matrix. The spatial attention calculation aims to highlight the spatial areas in the image that are most important for product identification and suppress insignificant areas. According to the spatial weight matrix, the multi-receptive field feature map is weightedly fused to obtain the second-level feature data. The second-level feature data is processed by the feature integration module. The feature integration module contains two fully connected layers and one Softmax layer. The function of the two fully connected layers is to further integrate and transform the extracted features, making the features more compact and discriminative. The Softmax layer is used to normalize the features and convert them into probability distributions to obtain the final feature data of medical products.

[0035] 103. Inputting the medical product feature data into the medical product identification model to perform medical product identification and obtain a medical product identification result;

[0036] Specifically, the feature data of medical products are input into the product recognition backbone network for feature extraction and recognition. The product recognition backbone network adopts the DenseNet structure. DenseNet (densely connected network) has high feature reusability. By implementing dense skip connections in the network, features are fully shared between layers, improving the efficiency of feature extraction and the training effect of the network. The network contains four densely connected blocks, each of which has six convolutional layers. The connection between the convolutional layers adopts dense skip connections. The output of each convolution layer will be passed to all subsequent layers, forming a feature sharing mechanism to obtain a dense map of product features. Multi-scale feature fusion operations are performed on the dense map of product features to unify features of different scales into the same dimension for subsequent recognition. The number of channels of feature maps of different scales is unified to 256 through a 1×1 convolution layer, which effectively reduces the computational complexity while maintaining the integrity of the features. The spatial dimensions of each feature map are adjusted to the same size through bilinear interpolation to obtain product feature fusion data. The fused product feature data is fed into the type recognition branch network for product type identification. The branch network consists of three parallel attention modules, each of which is composed of a channel attention unit and a spatial attention unit connected in series. The channel attention unit assigns weights to different channels in the feature map, highlighting important channels and suppressing unimportant features. The spatial attention unit, on the other hand, focuses on features at spatial locations, enhancing important regional features related to the product category. After processing by these three parallel attention modules, a product type feature vector is obtained. This product type feature vector is fed into the type classifier for classification and identification. The type classifier consists of two fully connected layers and one softmax layer. The output dimension of the first fully connected layer is set to 512 to reduce and integrate the product type features. The output dimension of the second fully connected layer is the same as the number of medical product categories, mapping the features into the specific product category space. The softmax layer converts the output results into a probability distribution for each category, ultimately obtaining the product type information. While identifying product types, the product location is also detected. The fused product feature data is fed into the location detection branch network for location identification. The location detection branch network uses a feature pyramid network (FPN) structure. The FPN structure effectively integrates multi-scale features through a top-down feature pyramid and lateral connections, resulting in a multi-level location feature map. Region proposals are calculated for the multi-level location feature map, and candidate region features are generated using three anchor box generators of different scales. Each anchor box generator is used to generate candidate regions of different sizes and proportions, ensuring that products can be effectively identified regardless of their size in the image. Non-maximum suppression is used to eliminate duplicate candidate regions and retain the most representative candidate region features. The dimension of each candidate region feature is set to 256.The candidate region features are input into the position regressor, which consists of four fully connected layers. Each layer uses the ReLU activation function to enhance the network's nonlinear representation and better fit the product's location features. The position regressor outputs four coordinate values, representing the position of the product's bounding box within the image—the specific location of the product. Label alignment is performed on the product type information and the product location information obtained from position detection. The label alignment module associates and matches the product type with its location information, resulting in a complete medical product recognition result.

[0037] 104. Perform inventory detection based on the medical product identification results, generate product inventory data, and send the product inventory data to the mobile management terminal to generate a replenishment task instruction;

[0038] Specifically, the product type and location information from the medical product identification results are input into the vending machine inventory detection unit. The product type information is classified and statistically processed to generate a product category quantity table showing the specific quantity of each product in the vending machine. Simultaneously, based on the product location information, a product aisle distribution map is generated, identifying the specific location of each product in the vending machine. The product category quantity table is compared with the product layout parameters in the initial product shelving plan. By comparing the current product inventory with the original plan, the missing products in the vending machine are identified, generating missing product data. Based on the missing product data, a replenishment demand matrix is ​​established to describe the replenishment demand for each missing product. The aisle distribution map and the product layout parameters are spatially mapped and analyzed. By comparing the actual product placement with the initial planned positions, product position deviation data is generated, indicating the deviation between the product placement in the vending machine and the planned plan. Based on this deviation data, aisle adjustment suggestions are generated to optimize the placement of products in the vending machine, ensuring that the products are more consistent with the original plan, improving customer accessibility, and improving the overall operational efficiency of the vending machine. The replenishment demand matrix is ​​prioritized and calculated to determine which products require replenishment first based on their urgency. To determine replenishment urgency, replenishment thresholds are set based on replenishment threshold parameters to generate graded replenishment demand data, categorizing replenishment needs for different products according to their urgency. This graded replenishment demand data is integrated with channel adjustment recommendations to generate comprehensive inventory data. To ensure the security of inventory data during transmission, the data encryption module encrypts the inventory data to generate encrypted inventory data. This encrypted inventory data is transmitted to a mobile management terminal via a wireless communication module. Upon receiving this data, the mobile management terminal performs identity authentication to ensure that only authorized personnel can access the inventory data. After identity authentication, the encrypted inventory data is decrypted to generate the terminal inventory data. A time-sharing replenishment plan is generated based on the terminal inventory data and sales period parameters. Based on sales performance during different time periods, a reasonable replenishment schedule is developed to ensure timely replenishment before peak sales periods, thereby maximizing customer satisfaction. Based on the time-sharing replenishment plan, the replenishment path is optimized so that the shortest and most efficient replenishment route can be selected when executing the replenishment task, thereby improving replenishment efficiency and reducing operating costs. The replenishment path data is input into the replenishment scheduling module, which combines the current staffing information to perform task allocation calculations. The replenishment tasks are rationally assigned to the available staff to ensure that the replenishment tasks can be completed efficiently. Through the above steps, the replenishment task instructions are generated, and the replenishment task content, replenishment time and replenishment route of each staff member are clearly defined, making the entire replenishment process automated and efficient, ensuring that the goods in the vending machine are always sufficient to meet medical needs.

[0039] 105. Input the RFID tag information and replenishment task instructions into the replenishment execution unit, generate replenishment operation data, and update the real-time inventory information;

[0040] Specifically, the replenishment path data and hierarchical replenishment demand data in the replenishment task instruction are input into the replenishment execution unit. Based on the replenishment path data, the replenishment execution unit performs task planning for each replenishment station, generating a replenishment execution task sequence. Based on the replenishment execution task sequence, an RFID reader is configured and scans the RFID tags of the replenishment items to obtain the tag data. The tag data of the replenishment items is matched and verified against the product category and quantity table to ensure that the actual replenished items are consistent with the planned items, generating a product verification result. After the product verification is completed, replenishment execution status data is generated based on the verification results. The replenishment execution status data records the execution status of each replenishment task, including information such as whether the replenishment was completed successfully and whether the items meet the requirements. To optimize the replenishment process, the replenishment execution status data is processed in conjunction with aisle adjustment suggestions to generate aisle placement guidance data. This data specifies the placement of each item within the vending machine, ensuring that the items are displayed according to the optimal layout. During the replenishment process, the replenishment progress is monitored in real time. By monitoring product verification results and replenishment execution status data, anomalies during the replenishment process are promptly detected and replenishment anomaly signals are generated. Replenishment anomaly signals include issues such as incorrect product quantities and tag reading failures. To process these anomaly signals, they are compared with replenishment threshold parameters to generate replenishment warning data. This replenishment warning data effectively alerts managers to issues with certain replenishment operations, requiring timely adjustments to prevent inventory impacts. This replenishment warning data is transmitted via wireless communication to a mobile management terminal. Upon receiving the warning data, the mobile management terminal classifies and processes it to generate a replenishment adjustment plan. Different response strategies are developed based on the severity of the warning. More urgent issues are prioritized, while minor deviations are corrected during the next replenishment. The replenishment adjustment plan ensures timely and appropriate adjustments to replenishment tasks when problems arise, ensuring the smooth progress of the replenishment process. Replenishment operation data is generated based on the replenishment adjustment plan and aisle placement guidance data, recording every replenishment step, including product placement and adjustment plan. This replenishment operation data is stored in the medical vending platform's database to create a replenishment history. Data fusion of replenishment history and product layout parameters is performed to update the medical vending platform's real-time inventory information. This fusion process enables the vending platform to adjust its inventory data based on the latest replenishment status, ensuring the accuracy and timeliness of inventory information. This updated real-time inventory information is fed back to the vending machine's inventory monitoring unit, which monitors and manages the status of products within the vending machine.

[0041] 106. Adjust the initial product listing plan based on real-time inventory information and historical sales data to generate a target product listing plan.

[0042] Specifically, real-time inventory information and replenishment history records are linked according to a product category quantity table to generate product flow data, reflecting the flow of various products into and out of the vending machine. This product flow data is analyzed over time to determine product sales over different time periods and generate product sales samples. Based on these product sales samples and product location deviation data, a sales location correlation matrix is ​​constructed. This sales location correlation matrix describes the relationship between product sales volume and its location within the vending machine, effectively identifying which locations have a significant impact on product sales. Cluster analysis is then performed on the sales location correlation matrix to generate product zone popularity data. This data reveals the sales activity of different zones within the vending machine, specifically which zones have a higher concentration of product sales during certain time periods. This data is cross-validated with tiered replenishment demand data to generate product turnover rate data, reflecting the speed and frequency of product sales. This allows us to determine whether certain products are slow-moving or hot-selling within a specific time period. Based on this turnover rate data, a product zoning configuration plan is generated. Optimize product display locations based on sales activity. For example, products with high turnover rates are placed in locations more accessible to customers to improve overall sales efficiency. Spatial mapping is performed between the product zoning plan and aisle layout guidance data to generate optimized layout data. This optimized layout data includes the optimal placement of products within the vending machine to maximize product display and increase sales opportunities. Product layout parameters are updated based on the optimized layout data. Simultaneously, replenishment threshold parameters are dynamically updated based on product turnover data to generate updated replenishment threshold parameters. The replenishment threshold parameters determine the safety stock level for each product within the vending machine. Dynamically updating the replenishment threshold parameters ensures timely replenishment of product inventory in vending machines during significant sales fluctuations, preventing stockouts or backlogs. The updated replenishment threshold parameters are input into the replenishment warning module, which issues alerts based on the new thresholds, allowing management to adjust replenishment tasks promptly. Time-period correlation analysis is performed based on the time-sharing replenishment plan and replenishment execution status data to generate time-period characteristic data. Time period characteristic data describes the sales characteristics of products within different time periods, such as significant increases or decreases in sales of certain products during certain time periods. Based on this time period characteristic data, sales period parameters are updated to more accurately reflect current sales patterns and trends. The updated product layout parameters, replenishment threshold parameters, and sales period parameters are subjected to a model fusion operation to obtain product sales characteristics, which include comprehensive information about the product layout within the vending machine, replenishment strategy, and sales patterns. Based on the product sales characteristics, the parameters of the initial product listing plan are updated to obtain the final target product listing plan.The target product listing plan is developed through multiple optimizations based on the initial plan, incorporating real-time inventory information, historical sales data, and product sales characteristics. This maximizes the vending machine's sales efficiency and meets customer purchasing needs. The target product listing plan is stored in the medical vending platform's operational database to ensure that the platform can operate according to the optimal product display and replenishment strategy in subsequent operations.

[0043] In the embodiment of the present invention, through deep learning and computer vision technology, in conjunction with the RFID identification system, accurate identification of medical products and real-time inventory monitoring are achieved; a two-stage feature extractor and residual learning method are adopted to effectively improve the accuracy of product identification in complex environments; by establishing a dynamic product shelving solution optimization mechanism, intelligent adjustment of product layout and automatic optimization of replenishment strategy are achieved; by introducing mobile management terminals and replenishment execution units, a complete replenishment-acceptance-shelf closed-loop management system is constructed, which significantly improves the operational efficiency and service quality of the medical vending platform; through intelligent analysis of real-time inventory information and sales data, an adaptive optimization mechanism for the sales model is established, which realizes the rational allocation of product resources and effective control of operating costs.

[0044] In a specific embodiment, the process of executing step 101 may specifically include the following steps:

[0045] Clean the sales records of each vending machine in the medical vending platform to obtain valid sales data, and classify the valid sales data according to product category and sales time to obtain classified sales data;

[0046] Perform time series analysis on classified sales data to obtain sales curves for each period. Calculate the sales frequency of goods in each period based on the sales curves to obtain sales period parameters.

[0047] Input sales period parameters into the commodity turnover rate calculation model to obtain commodity turnover speed data. According to the commodity turnover speed data, the safety stock amount of each commodity is set to obtain the replenishment threshold parameter.

[0048] Perform spatial distribution analysis on the product placement data of each vending machine to obtain product location popularity data, and prioritize the product placement locations based on the product location popularity data to obtain location priority data;

[0049] Perform correlation analysis on location priority data and sales period parameters to obtain location and period matching data. Generate a product location distribution matrix based on the matching data to obtain product layout parameters.

[0050] Generate an initial product placement plan based on replenishment threshold parameters, product layout parameters, and sales period parameters. Based on the initial product placement plan, monitor the remaining quantity of each product in the vending machine to obtain the product out-of-stock risk value. Determine the priority of image acquisition based on the product out-of-stock risk value.

[0051] The merchandise display in the vending machine is spatially mapped according to the merchandise layout parameters to obtain merchandise coordinate data. The acquisition angle of the image acquisition device is set based on the merchandise coordinate data. According to the image acquisition angle and the priority of image acquisition, the image acquisition device is controlled to perform adaptive angle acquisition to obtain medical merchandise image data.

[0052] Specifically, incomplete, unreasonable or duplicate data is removed from the sales records of the vending machine, and the data obtained can truly reflect the actual sales situation. For example, some repeated transactions recorded due to vending machine failures and erroneous records caused by network delays need to be eliminated. After this process, valid sales data is obtained. The valid sales data is classified according to product category and sales time to obtain classified sales data. Data classification helps to understand the sales patterns of different categories of products in different time periods. Time series analysis is performed on the classified sales data. By modeling the sales situation in different time periods, the sales curve of each product in each time period is obtained. Assume that the function of the sales volume of the product changing over time is S i (t), where S i (t) represents the sales volume of the i-th category of goods at time t. Analyze the sales curve and calculate the sales frequency of the goods in each period, denoted as f i (t), the formula is:

[0053]

[0054] Among them, S i，mean It represents the average sales volume of the i-th category of goods in the entire time period, and T is the total length of the time period. By calculating the sales frequency f i (t), we can get the sales activity of the product in each time period, and thus get the sales period parameter, which reflects the demand fluctuation of the product in a specific time period. The sales period parameter is input into the product turnover rate calculation model. The product turnover rate is used to measure the speed at which the product is sold from the shelf. The calculation formula of the product turnover rate is:

[0055]

[0056] Among them, β i Denotes the turnover rate of the i-th category of goods, D i Indicates the total sales volume of the product in a specific period of time, I i Represents the average inventory of goods. By calculating the product turnover rate β i, get the turnover speed data of the goods. According to the turnover speed, set the safety stock of each type of goods, recorded as Q i , and its calculation formula is:

[0057] Q i =β i ×T s ;

[0058] Among them, T s Represents the replenishment cycle. Safety stock Q i Ensure that each product has sufficient inventory during the replenishment cycle to cope with demand fluctuations and obtain the replenishment threshold parameter. In order to optimize the layout of the vending machines, the spatial distribution of the product placement data of each vending machine is analyzed to obtain the product location popularity data. The location popularity data reflects the popularity of different locations in the vending machine and is calculated by recording the frequency of customers purchasing products at the location. Assume that the popularity of a certain location is H j , the formula is:

[0059] H j =∑ t p j (t);

[0060] Among them, p j (t) represents the number of purchases at the jth position in the vending machine at time t. j , prioritize the positions of the products in the vending machine and obtain the position priority data. The position priority data is correlated with the sales period parameters to calculate the matching degree of different products in different positions and time periods, and generate the matching degree data of position and time period. The matching degree is measured by the correlation coefficient. Assume that the matching degree of a certain product i at position j is ω ij , and its calculation formula is:

[0061]

[0062] in, and Represent the average sales frequency of product i and the average popularity of location j, respectively. Using these matching data, we generate a product location distribution matrix and obtain product layout parameters. Based on the replenishment threshold parameters, product layout parameters, and sales period parameters, we generate an initial product placement plan, describing the optimal placement and required inventory for each product in the vending machine. Based on the initial product placement plan, we monitor the remaining quantity of each product in the vending machine and obtain the product's out-of-stock risk value. Out-of-stock risk value κ 缺，i The calculation formula is:

[0063]

[0064] Among them, Bi Represents the current inventory. According to the out-of-stock risk value κ 缺，i , determine the products that need to be captured first so that they can be replenished in time. According to the product layout parameters, the display of the products in the vending machine is spatially mapped to obtain the coordinate data of each product. Assume that the position coordinates of product i in the vending machine are (x i ,y i ), these coordinate data help determine the acquisition angle θ of the image acquisition device i The acquisition angle is calculated based on the location of the product and the camera installation position. The formula is:

[0065]

[0066] Among them, (x c ,y c ) are the coordinates of the camera. Based on the image acquisition angle and the image acquisition priority, the image acquisition device is controlled to perform adaptive angle acquisition to obtain image data of the medical product.

[0067] In a specific embodiment, the process of executing step 102 may specifically include the following steps:

[0068] Performing brightness analysis on the medical product image data to obtain an image brightness distribution matrix, and performing block processing on the medical product image data according to the image brightness distribution matrix to obtain an image block sequence;

[0069] The image block sequence is input into the backbone feature extraction layer of the residual network. Feature extraction is performed through three residual units. Each residual unit contains two 3×3 convolutional layers and a short-circuit connection. The convolutional layer uses the ReLU activation function and batch normalization to obtain the backbone feature map.

[0070] Perform residual mapping calculation on the backbone feature map to obtain a residual feature map, and perform element-level sum operation on the residual feature map and the image block sequence to obtain enhanced image data;

[0071] The enhanced image data is input into the first-level feature extraction network of the two-level feature extractor. The first-level feature extraction network consists of a pyramid structure consisting of five convolutional layers. After each convolution layer, a maximum pooling operation is used to obtain a multi-scale feature map.

[0072] Perform channel attention calculation on the multi-scale feature map to obtain the channel weight matrix, and perform weighted processing on the multi-scale feature map according to the channel weight matrix to obtain the first-level feature data;

[0073] The first-level feature data is input into the second-level feature extraction network of the two-level feature extractor. The second-level feature extraction network adopts four parallel dilated convolution branches with dilation rates set to 1, 2, 4, and 8 respectively to obtain multi-receptive field feature maps.

[0074] Perform spatial attention calculation on the multi-receptive field feature map to obtain the spatial weight matrix, and perform weighted fusion on the multi-receptive field feature map according to the spatial weight matrix to obtain the second-level feature data;

[0075] The second-level feature data is processed by the feature integration module, which contains two fully connected layers and one Softmax layer to obtain the feature data of medical products.

[0076] Specifically, the brightness of the medical product image data is analyzed to obtain the image brightness distribution matrix. The image brightness distribution matrix is ​​used to describe the brightness value of each pixel in the image, forming a two-dimensional matrix. Assuming that the input image is λ, its size is ξ×τ, where ξ and τ represent the height and width of the image respectively, then each element L of the brightness distribution matrix L ij Calculated by the following formula:

[0077] L ij =0.299×R ij +0.587×G ij +0.114×B ij ;

[0078] Among them, R ij , G ij 、B ij Represent the pixel values ​​of the red, green and blue channels of the image at position (i, j) respectively. Through the above formula, the brightness distribution matrix L reflects the brightness information of each position in the image, so that the image is divided into blocks according to these brightness values. The original image is divided into multiple smaller areas so that more detailed feature extraction can be performed on these areas. Assuming that the image λ is divided into image blocks of size P×P, the image block sequence S={S1,S2,...,S K}, where K represents the number of image blocks. Each image block S k Containing P×P pixels, these image blocks serve as independent inputs to facilitate subsequent feature extraction. The image block sequence is input into the backbone feature extraction layer of the residual network. The backbone feature extraction layer of the residual network adopts a structure composed of three residual units, each of which contains two 3×3 convolutional layers and a short-circuit connection. The function of the convolutional layer is to extract features from the input image blocks. The convolution kernel size of each convolutional layer is 3×3, and its output feature map is represented as:

[0079] F1=ReLU(BN(W1*F 1-1+b1);

[0080] Among them, F l-1 is the input feature map of the previous layer, W l represents the weight of the current convolutional layer, b l is the bias term, * represents the convolution operation, BN represents batch normalization, which is used to improve the training stability and convergence speed of the model, and ReLU represents the activation function, which is used to increase the nonlinear expression ability of the network. The role of the short-circuit connection is to add the input directly to the output of the convolution layer to form a skip connection, so that the gradient can be better transmitted and avoid the gradient vanishing problem in the deep network. After processing by three residual units, the backbone feature map F is obtained. main . For the backbone feature map F main Perform residual mapping calculation to obtain the residual feature map F res The purpose of residual mapping is to capture subtle changes in the feature map and enhance the model’s ability to learn local details. res Perform element-level addition operation with the original image block sequence S to obtain enhanced image data F enh , and its calculation formula is:

[0081] F enh =F res +S;

[0082] The information of the original image is fused with the extracted residual features to enhance the image’s feature representation ability. The enhanced image data is input into the first-level feature extraction network of the two-level feature extractor. The first-level feature extraction network adopts a pyramid structure consisting of five convolutional layers. After each convolution layer, a maximum pooling operation is used to reduce the spatial size of the feature map and enhance the network’s perception of features at different scales. Assume that the input feature map of the first-level network is F enh , the output after each layer of convolution is:

[0083] F (i) =MaxPool(ReLU(BN(W (i) *F (i-1) +b (i) )));

[0084] Among them, F (i) is the output feature map of the i-th layer convolution, W (i) and b (i) are the weight and bias of the i-th layer of convolution respectively. After five layers of convolution and maximum pooling, the multi-scale feature map F is obtained. multi , these feature maps contain product information at different scales, which helps to capture the global and local features of the product. multiPerform channel attention calculation to highlight the most valuable channels in the feature map and suppress irrelevant features. The calculation formula of the channel attention weight matrix C is:

[0085] C=σ(W c F multi +b c );

[0086] Among them, W c and b c are weights and biases respectively, and σ represents the activation function (usually Sigmoid), which is used to map the output to the interval (0, 1). The multi-scale feature map is weighted according to the channel weight matrix to obtain the first-level feature data F level1 :

[0087] F level1 =C·F multi

[0088] The first-level feature data is input into the second-level feature extraction network of the two-level feature extractor. The second-level feature extraction network uses four parallel dilated convolution branches, with dilated rates set to 1, 2, 4, and 8 respectively. The purpose of dilated convolution is to increase the receptive field without increasing the parameters, thereby capturing features in a wider range. Let the output of the i-th dilated convolution be F dil,i , and its calculation formula is:

[0089]

[0090] Among them, *r i The void rate is r i The convolution operation, W dil,i and b dil，i are the weight and bias of the i-th dilated convolution respectively. Through four parallel dilated convolution branches, the multi-receptive field feature map F is obtained. multi-dil . For multiple receptive field feature maps F multi-dil Perform spatial attention calculation to highlight the most important spatial positions in the feature map. The calculation formula of the spatial attention weight matrix is:

[0091] ε=σ(W s *F multi-dil +b s );

[0092] The multi-receptive field feature maps are weighted fused according to the spatial weight matrix to obtain the second-level feature data F level2 :

[0093] F level2 =ε·F multi-dil ;

[0094] The second-level feature data F level2 The feature integration module consists of two fully connected layers and one Softmax layer. The first fully connected layer is used to reduce the dimension of the feature and obtain the intermediate feature F fc1 :

[0095] F fc1 =W fc1 F level2 +b fc1 ;

[0096] Among them, W fc1 and b fc1 are the weights and biases of the first fully connected layer. The output dimension of the second fully connected layer is the same as the number of medical product categories and is used for classification prediction. The output is F fc2 :

[0097] F fc2 =W fc2 F fc1 +b fc2 ;

[0098] Through the Softmax layer, F fc2 Normalization is performed to obtain the probability of each product category and the characteristic data of medical products. The calculation formula of Softmax is:

[0099]

[0100] Among them, P i It represents the probability that the product belongs to the i-th category.

[0101] In a specific embodiment, the process of executing step 103 may specifically include the following steps:

[0102] The medical product feature data is input into the product recognition backbone network. The product recognition backbone network adopts the DenseNet structure, which contains four densely connected blocks. Each densely connected block has six convolutional layers. Dense skip connections are used between adjacent convolutional layers to obtain a dense map of product features.

[0103] Perform multi-scale feature fusion on the dense product feature map. Use a 1×1 convolutional layer to unify the number of channels of feature maps of different scales to 256. Use bilinear interpolation to adjust the spatial size of the feature maps to the same size to obtain product feature fusion data.

[0104] The product feature fusion data is input into the type recognition branch network. The type recognition branch network contains three parallel attention modules. Each attention module is composed of a channel attention unit and a spatial attention unit in series to obtain the product type feature vector.

[0105] The product type feature vector is input into the type classifier, which consists of two fully connected layers and a softmax layer. The output dimension of the first fully connected layer is 512, and the output dimension of the second fully connected layer is the same as the number of medical product categories, to obtain product type information.

[0106] The product feature fusion data is input into the location detection branch network. The location detection branch network adopts the FPN structure, which includes a top-down feature pyramid and horizontal connections to obtain a multi-level location feature map.

[0107] Perform region proposal calculation on the multi-level position feature map, and obtain candidate region features through three anchor box generators of different scales and non-maximum suppression operations. The dimension of each candidate region feature is 256;

[0108] The candidate region features are input into the position regressor. The position regressor consists of four fully connected layers. Each layer uses the ReLU activation function. The last layer outputs four coordinate values ​​representing the bounding box coordinates of the product location to obtain the product location information.

[0109] The product type information and product location information are label-aligned, and the product type and location information are associated and matched through the label alignment module to obtain the medical product identification result.

[0110] Specifically, the product recognition backbone network is used to perform deep feature extraction on the input medical product feature data. The product recognition backbone network adopts the DenseNet (densely connected network) structure, which contains multiple densely connected blocks. The network consists of four densely connected blocks, each of which contains six convolutional layers, and dense jump connections are used between adjacent convolutional layers. This dense connection method can effectively improve the reusability of features and enhance the expression ability of the network. For the input medical product feature data F in , after being processed by the first dense connection block, the output feature representation is:

[0111] F u =Concat(F in ,f1(F in ), f2(f1(F in )),...,f6(f5(…f1(F in ))));

[0112] Among them, f i (·) represents the convolution operation of the i-th layer, and Concat represents the feature concatenation operation. In this way, the output of each convolution layer is concatenated with the output of all previous layers to form an output with richer features. After four such densely connected blocks, the feature dense map F of the product is obtained. denseThis feature-intensive map contains detailed feature descriptions of medical products at different levels and different spatial locations. dense Perform multi-scale feature fusion operations to integrate feature information from different scales and enhance the descriptive ability of features. The number of channels of feature maps of different scales is unified to 256 through the 1×1 convolution layer to reduce the computational complexity while ensuring full expression of information. Assume that the feature map after 1×1 convolution is F conv , and its calculation formula is:

[0113] F conv =W conv *F dense +b conv ;

[0114] Among them, W conv represents the weight of the convolution kernel, b conv is the bias term, and * is the convolution operation. The bilinear interpolation method is used to adjust the spatial size of the feature map to the same size to obtain the product feature fusion data F fused The role of bilinear interpolation is to spatially align features of different scales for subsequent feature integration and processing. The obtained product feature fusion data F fused The input is fed into the category recognition branch network for product category identification. The category recognition branch network contains three parallel attention modules, each of which consists of a channel attention unit and a spatial attention unit in series. The channel attention unit weights each channel in the feature map to highlight important channel features. The channel weight matrix Ω is calculated as:

[0115] Ω=σ(W c F fused +b c );

[0116] Among them, W c is the weight of the channel weight matrix, b c is the bias, and σ represents the activation function (usually the Sigmoid function), which is used to limit the weight value to the interval (0, 1). The spatial attention unit is used to calculate the weight of each spatial position to highlight the most valuable spatial position in the feature map. Spatial weight matrix The calculation formula is:

[0117]

[0118] Through the processing of channel attention and spatial attention, the type recognition branch network obtains the product type feature vector F type , contains the category-related features of medical products. The product type feature vector F typeThe input is sent to the type classifier to classify the product. The type classifier consists of two fully connected layers and a Softmax layer. The output dimension of the first fully connected layer is 512, which is used to reduce and integrate the product type features. Output F fd1 Expressed as:

[0119] F fd1 =W fd1 F type +b fd1

[0120] Among them, W fd1 and b fd1 are the weights and biases of the first fully connected layer respectively. The output dimension of the second fully connected layer is the same as the number of medical product categories, which is used to map the features to the specific product category space, and the output is F fd2 :

[0121] F fd2 =W fd2 F fd1 +b fd2 ;

[0122] Through the Softmax layer, F fd2 Perform normalization processing to obtain the probability distribution of each product category and obtain the product type information η i :

[0123]

[0124] Among them, η i Indicates the probability that the product belongs to the i-th category. At the same time, the product features are fused with data F fused Input to the position detection branch network, the position detection branch network adopts FPN (feature pyramid network) structure, the FPN structure contains top-down feature pyramid and lateral connection, which is used to generate multi-level position feature map F loc The advantage of FPN is that it can simultaneously utilize the semantic information of high-level features and the spatial details of low-level features to achieve more accurate position detection. loc Perform region proposal calculations and generate candidate regions using three anchor box generators of different scales. Each anchor box generator is used to generate candidate regions of different sizes and proportions to ensure that products can be effectively detected regardless of their size in the image. Non-maximum suppression is used to remove duplicate and redundant candidate regions and retain the most representative candidate region features ρ. k , where k represents the index of the candidate region. The dimension of each candidate region feature is 256, which is used to describe the characteristic information of the product in the region. kThe input is fed into the position regressor, which consists of four fully connected layers, each using a ReLU activation function to enhance the network's nonlinear representation capabilities. After processing through the four fully connected layers, the position regressor outputs four coordinate values ​​(x, y, w, h), representing the bounding box coordinates of the product's location, where x and y represent the center of the bounding box, and w and h represent the width and height of the bounding box, respectively. These coordinate values ​​constitute the product's location information and are used to accurately locate the product in the image. The product type information and product location information are label-aligned, and the label alignment module associates and matches the product type with the location information to obtain the final medical product recognition result.

[0125] In a specific embodiment, the process of executing step 104 may specifically include the following steps:

[0126] The product type information and product location information in the medical product identification results are input into the vending machine inventory detection unit, the product type information is classified and statistically processed to obtain a product category quantity table, and a product aisle distribution map is generated based on the product location information;

[0127] Compare the product category quantity table with the product layout parameters in the initial product listing plan to obtain missing product data, and establish a replenishment demand matrix based on the missing product data;

[0128] Perform spatial mapping analysis on the product aisle distribution map and product layout parameters to obtain product location deviation data, and generate aisle adjustment suggestions based on the product location deviation data;

[0129] Prioritize and calculate the replenishment demand matrix, set the replenishment level threshold based on the replenishment threshold parameter, and obtain graded replenishment demand data;

[0130] Integrate and process the hierarchical replenishment demand data and the channel adjustment suggestions to obtain product inventory data, and encrypt the product inventory data through the data encryption module to obtain encrypted inventory data;

[0131] The encrypted inventory data is sent to the mobile management terminal through the wireless communication module, and identity authentication and data decryption are performed on the mobile management terminal to obtain the terminal inventory data;

[0132] Generate a time-sharing replenishment plan based on terminal inventory data and sales period parameters, and optimize the replenishment path based on the time-sharing replenishment plan to obtain replenishment path data. The replenishment path data is input into the replenishment scheduling module, and task allocation calculation is performed in combination with personnel configuration information to obtain replenishment task instructions.

[0133] Specifically, the product type information is classified and statistically processed to generate a product category quantity table. The product type information is output by the product recognition model and contains the specific category of each product in the vending machine, such as masks, disinfectants, thermometers, etc. By classifying and statistically analyzing the product type information, the quantity of each product is obtained, which is recorded as N i , where i represents the category of the product, N i Represents the number of products in category i. The product category quantity table A is represented as a vector:

[0134] A=[N1,N2,...,N m ];

[0135] Where m is the total number of different product categories in the vending machine. This quantity table intuitively shows the inventory status of various products in the vending machine. At the same time, based on the product location information in the medical product recognition results, a product aisle distribution map is generated. Assume that the location of each aisle in the vending machine is (x j ,y j ), the location of each product is represented by its center coordinates. By integrating these location information, we can get the product aisle distribution map P, which describes the specific placement of each product in the vending machine. init Perform a comparison operation to detect whether there is any missing product in the vending machine. Product layout parameter L init Describe the expected quantity of each product when it is initially put on the shelf, denoted as L init =[L1, L2, ..., L m ]. Missing data for products are calculated using the following formula:

[0136] α i =L i -N i , for i=1,2,...,m;

[0137] Among them, α i Indicates the number of missing items in category i. i If α > 0, it means that the inventory of the product is insufficient and needs to be replenished. m ], establish the replenishment demand matrix R, which describes the replenishment demand of each commodity category. Combine the commodity channel distribution map P with the commodity layout parameter L init Perform spatial mapping analysis to obtain product location deviation data. Product location deviation data is used to describe the difference between the actual product placement location and the initial planned location. Suppose the expected location of a product is Its actual position is (x i ,y i ), then the position deviation Δi Expressed as:

[0138]

[0139] By calculating the position deviation of all commodities, we can get the commodity position deviation data matrix Δ=[Δ1,Δ2,...,Δ m Based on the position deviation data, aisle adjustment suggestions are generated to optimize the placement of goods in the vending machine, ensuring that the product location is more reasonable to improve customer accessibility and purchase convenience. Prioritize the replenishment demand matrix R to determine which products need to be replenished first. Based on the replenishment threshold parameter Set the replenishment level threshold V i Determine the urgency of replenishment. Assume that the replenishment threshold parameter is The replenishment level is expressed as:

[0140]

[0141] Among them, V i =1 means that the products of category i need to be replenished first, V i =0 means that the product does not need urgent replenishment. According to the replenishment level, the graded replenishment demand data V = [V1, V2, ..., V m ] Integrate the graded replenishment demand data V and the channel adjustment suggestions to obtain the product inventory data The data includes the current inventory quantity, replenishment requirements and placement adjustment suggestions of each product. The data encryption module is used to encrypt the product inventory data to obtain the encrypted inventory data S enc Protect the privacy and integrity of data and avoid attacks or tampering during transmission. Encrypted inventory data S enc After receiving the data, the mobile management terminal will perform identity authentication to ensure that only authorized personnel can access the sensitive information. enc Decrypt and obtain the terminal inventory data S dec , used for subsequent replenishment management and decision-making. According to the terminal inventory data S dec and sales period parameter T sale , generate a time-sharing replenishment plan. The sales period parameter describes the sales of each product in different time periods, such as the sales peaks in the morning and at noon. The goal of the time-sharing replenishment plan is to reasonably arrange the replenishment time according to the sales peaks and troughs to maximize customer demand and reduce the risk of product out-of-stock. The time-sharing replenishment plan is represented by a matrix P time , where each element represents the replenishment arrangement of a certain commodity in a certain period of time. time, optimize the replenishment path and obtain the replenishment path data R path The purpose of replenishment path optimization is to find the optimal replenishment route to reduce replenishment time and operating costs. The optimization of replenishment path is achieved by solving the traveling salesman problem, where the objective function of path length is:

[0142]

[0143] Among them, d(p i , p i+1 ) represents the replenishment site p i and p i+1 The distance between them, n is the number of replenishment stations. By minimizing the path length Find the optimal replenishment path. path This is input into the replenishment scheduling module, which, combined with the current staffing information U, performs task allocation calculations to generate replenishment task instructions Ψ. The replenishment task instructions Ψ contain the specific tasks for each replenishment staff member, including detailed information such as the type of goods to be replenished, the quantity, the replenishment time, and the replenishment route.

[0144] In a specific embodiment, the process of executing step 105 may specifically include the following steps:

[0145] Input the replenishment path data and hierarchical replenishment demand data in the replenishment task instruction into the replenishment execution unit, perform task planning for the replenishment sites based on the replenishment path data, and obtain the replenishment execution task sequence;

[0146] The RFID reader is configured based on the replenishment execution task sequence, and the RFID tag information of the replenishment product is scanned by the RFID reader to obtain the replenishment product tag data;

[0147] Match and verify the replenishment product label data with the product category and quantity table to obtain the product verification results, and generate replenishment execution status data based on the product verification results;

[0148] Collaboratively process replenishment execution status data and aisle adjustment suggestions to obtain aisle placement guidance data, and then spatially arrange replenishment products based on the aisle placement guidance data;

[0149] Monitor replenishment progress in real time based on product verification results and replenishment execution status data, obtain replenishment anomaly signals, and compare the replenishment anomaly signals with replenishment threshold parameters to generate replenishment warning data;

[0150] Send replenishment warning data to the mobile management terminal through the wireless communication module, perform hierarchical processing on the replenishment warning data, and obtain a replenishment adjustment plan;

[0151] Generate replenishment operation data based on the replenishment adjustment plan and the cargo lane layout guidance data, and write the replenishment operation data into the database of the medical vending platform to obtain the replenishment history record;

[0152] Perform data fusion processing on replenishment history records and product layout parameters, update the real-time inventory information of the medical vending platform, and feed back the real-time inventory information to the vending machine inventory detection unit.

[0153] Specifically, the replenishment path data and the hierarchical replenishment demand data in the replenishment task instructions are combined to plan the specific replenishment operations. The replenishment path data describes the order of visiting each replenishment station, while the hierarchical replenishment demand data contains the types and quantities of goods that need to be replenished at each station. Based on this information, the replenishment execution unit formulates the replenishment execution task sequence. Let P path Represents replenishment path data, where P path =[p1, p2, ..., p n ], represents the n replenishment sites that need to be visited in sequence. Let D level Represents hierarchical replenishment demand data, including the types and quantities of goods required to be replenished at each replenishment site. path and D level Combined, we get the replenishment execution task sequence S task , which has the form:

[0154]

[0155] in, Indicates that at the replenishment site p i The types and quantities of goods that need to be replenished are configured based on the replenishment execution task sequence to ensure that the RFID reader can accurately identify the tag information of the replenishment goods. The RFID reader scans the RFID tag information of each replenishment product to obtain the replenishment product tag data. Assume T rfid =[t1, t2, ..., t m ], where t i Represents the label information of the i-th replenishment product. These label information contains the unique identifier of each product, which is used to match and verify the product information in the system. rfid Verify the matching with the commodity category quantity table to ensure that the types and quantities of the replenished commodities are consistent with the demand. Assume that the commodity category quantity table is A = [N1, N2, ..., N m ], represents the expected inventory quantity of each commodity. rfid Match with A to get the product verification result V check , which has the form:

[0156] V check=[v1, v2, ..., v m ];

[0157] Among them, v i Indicates the verification result of the i-th category of goods, v i =1 means the verification is passed, v i =0 means verification failed. If some products fail to pass the verification, it means that there is an error in the replenishment process, such as the wrong type of replenishment product or insufficient quantity. Based on the product verification results, the replenishment execution status data S is generated. exec , used to record the execution status of replenishment tasks. exec Collaborate with the cargo lane adjustment suggestions to generate cargo lane layout guidance data G guide , which contains the specific placement location and aisle number of each product in the vending machine. According to the aisle placement guidance data, the replenishment products are spatially arranged and each product is placed in the vending machine according to the designated location. During the replenishment task execution, the replenishment progress is monitored in real time to detect and handle any abnormal situations in a timely manner. According to the product verification results V check and replenishment execution status data S exec , get the replenishment abnormal signal A alarm The replenishment abnormality signal is used to prompt the system of any abnormality in the replenishment process, such as insufficient quantity of goods, wrong type of goods or replenishment delay, etc. The replenishment abnormality signal is combined with the replenishment threshold parameter Q threshold Compare and generate replenishment warning data W warn , to determine which exceptions need to be handled immediately and which can be resolved later. The formula for replenishment warning is:

[0158]

[0159] Among them, W warn =1 indicates an exception that needs to be handled immediately, W warn=0 means it can be processed later. The replenishment warning data is sent to the mobile management terminal through the wireless communication module. After receiving these data, the mobile management terminal performs graded processing on the warning data to obtain a replenishment adjustment plan. Different response strategies are formulated according to the severity of the warning. More urgent problems are solved first, while minor deviations are corrected at the next replenishment. Based on the replenishment adjustment plan and the aisle layout guidance data, the final replenishment operation data is generated. The replenishment operation data includes every step of replenishment, the placement of the goods, and any necessary adjustments. The replenishment operation data is written into the database of the medical vending platform to generate a replenishment history record. The replenishment history record provides a detailed operation track for the vending platform for subsequent inventory management and analysis, helping to optimize the replenishment strategy. The replenishment history record is fused with the product layout parameters to update the real-time inventory information of the medical vending platform. Let the product layout parameter be L layout , the replenishment history is H history , through data fusion to obtain updated real-time inventory information I real :

[0160] I real =f fuse (H history , L layout );

[0161] Among them, f fuse Represents data fusion operation. Real-time inventory information I real This information includes the current inventory quantity, location, and replenishment history of each product. The updated real-time inventory information is fed back to the vending machine inventory detection unit so that the system can understand the inventory status of the products in the vending machine in real time and make timely replenishments and adjustments.

[0162] In a specific embodiment, the process of executing step 106 may specifically include the following steps:

[0163] Real-time inventory information and replenishment history records are linked according to the commodity category quantity table to obtain commodity flow data, and the commodity flow data is analyzed in the time dimension to obtain commodity sales samples;

[0164] Based on the product sales samples and product location deviation data, a sales location correlation matrix is ​​constructed, and cluster analysis is performed on the sales location correlation matrix to obtain product regional popularity data;

[0165] Cross-validate product area popularity data with tiered replenishment demand data to obtain product turnover rate data, and generate product zoning configuration plans based on the product turnover rate data;

[0166] Perform spatial mapping calculations on the product partition configuration plan and aisle layout guidance data to obtain optimized layout data, and update the product layout parameters based on the optimized layout data;

[0167] Dynamically update and calculate the replenishment threshold parameters based on the product sales rate data to obtain updated replenishment threshold parameters, and input the updated replenishment threshold parameters into the replenishment warning module;

[0168] Perform time period correlation analysis based on the time-sharing replenishment plan and replenishment execution status data to obtain time period feature data, and update sales time period parameters based on the time period feature data;

[0169] Performing pattern fusion operations on the updated product layout parameters, the updated replenishment threshold parameters, and the updated sales period parameters to obtain product sales features;

[0170] The parameters of the initial product listing plan are updated based on the product sales characteristics to obtain a target product listing plan, and the target product listing plan is written into the operation database of the medical vending platform.

[0171] Specifically, real-time inventory information and replenishment history records are integrated to form a complete record of commodity circulation. real Describes the current quantity of each product in the vending machine, and the replenishment history H history Contains detailed information for each replenishment operation. Product Category Quantity Table C type Used to define the category of each product, by integrating real-time inventory information real With replenishment history H history Perform data association to obtain commodity flow data F flow . F flow Record the changes of each category of goods over a period of time, including the increase (replenishment) and decrease (sales) of inventory. Assume that the turnover data of a certain category of goods i is F i (t), which is expressed as:

[0172] F i (t) = I real，i (t)-I real，i (t-1)+H history，i (t);

[0173] Among them, I real，i (t) represents the inventory of commodity i at time t, H history，i(t) represents the replenishment quantity at time t. By analyzing the time dimension of commodity flow data, commodity sales samples are obtained, which include the sales changes of various commodities in different time periods. The sales position association matrix is ​​constructed based on the commodity sales samples and commodity position deviation data. The commodity position deviation data describes the deviation between the position of each commodity in the vending machine and the expected position, which can help analyze whether the location of the commodity affects its sales. The element M of the sales position association matrix ij It represents the correlation between the sales of product category i at location j and its location. The calculation formula is:

[0174]

[0175] Among them, S i (t) represents the sales volume of commodity i at time t, is the average sales volume of commodity i, Δ j represents the deviation value of position j, is the average value of the position deviation. By performing cluster analysis on the sales position association matrix, we can obtain product area popularity data, which represents the sales activity of different areas in the vending machine, thereby identifying the impact of product placement on its sales. The product area popularity data is cross-validated with the graded replenishment demand data to obtain product turnover rate data. The product turnover rate reflects the sales speed and inventory turnover of the product, and is used to determine which products are selling well and which products are slow to sell. i The calculation formula is:

[0176]

[0177] Among them, S i (t) represents the sales volume of commodity i, I avg，i is the average inventory of product i. By calculating the turnover rate data, a product partition configuration plan is generated, which is used to re-plan the placement area of ​​the products, for example, placing products with higher turnover rates in locations that are easier for customers to see and use. A spatial mapping operation is performed on the product partition configuration plan and the aisle layout guidance data to obtain the optimized layout data. The optimized layout data contains the optimal placement position for each product in the vending machine. The spatial mapping operation combines the partition configuration plan with the aisle layout data to ensure that the placement of the products is in line with the sales rules and can maximize the use of the space resources of the vending machine. Based on the optimized layout data, the product layout parameters are updated to make the placement of the products in the vending machine more scientific and reasonable. At the same time, the replenishment threshold parameters are dynamically updated and calculated based on the product turnover rate data to obtain the updated replenishment threshold parameter Q newThe replenishment threshold parameter determines when a certain type of product needs to be replenished. By analyzing the turnover rate, if a certain type of product has a high turnover rate, its replenishment threshold is lowered accordingly to reduce the risk of out-of-stock; conversely, if a certain type of product has a low turnover rate, its replenishment threshold is appropriately increased to avoid inventory backlogs. The replenishment threshold update formula is:

[0178]

[0179] Among them, Q old，i is the old replenishment threshold of product i, is the adjustment coefficient, R avg is the average sales rate of all products. Through this formula, the replenishment threshold is dynamically adjusted to make it more in line with the actual sales situation of the product. The updated replenishment threshold parameters are input into the replenishment warning module so that a warning can be issued in time when the inventory is insufficient. According to the time-sharing replenishment plan and replenishment execution status data, a time period association analysis is performed to obtain the time period feature data. The time period feature data describes the sales characteristics of different products in each time period. For example, the sales volume of certain products increases or decreases significantly in a specific time period. Based on the time period feature data, the sales time period parameters are updated to optimize the replenishment and sales strategies of the vending machine to better meet customer needs. The updated product layout parameters, the updated replenishment threshold parameters and the updated sales time period parameters are subjected to pattern fusion operation to obtain the product sales feature F sales The pattern fusion operation combines feature information from different sources to generate a global product sales feature description, which is used to guide product listing and replenishment decisions. Product sales feature F sales It includes the product's sales situation, location layout, replenishment strategy and time characteristics, which is a comprehensive description of the product's sales performance. According to the product sales characteristics, the parameters of the initial product listing plan are updated to obtain the target product listing plan L target The target product listing plan is the result of optimization based on the initial plan, combined with real-time inventory, sales samples, sales rate, location popularity and other factors, which can maximize customer satisfaction and improve the operating efficiency of the vending machine. target Update using the following formula:

[0180] L target，i =L init,i +z·F sales,i

[0181] Among them, L init，i represents the parameter of product i in the initial listing plan, and z is the adjustment coefficient, which is used to control the impact of sales characteristics on the final listing plan. targetThe data is written into the operation database of the medical vending platform to ensure that the vending machine can operate according to the optimal product display and replenishment strategy in subsequent operations.

[0182] The above describes the mobile management method of the medical vending platform in the embodiment of the present invention. The following describes the mobile management device of the medical vending platform in the embodiment of the present invention. Figure 2 In one embodiment of the present invention, a mobile management device for a medical vending platform includes:

[0183] The acquisition module 201 is used to create an initial product shelf plan based on the vending machine operation data of the medical vending platform and to collect medical product image data;

[0184] Extraction module 202, for performing residual learning and two-level feature extraction on medical product image data to obtain medical product feature data;

[0185] Identification module 203, configured to input the medical commodity feature data into a medical commodity identification model to perform medical commodity identification and obtain a medical commodity identification result;

[0186] The detection module 204 is used to perform inventory detection based on the medical product identification results, generate product inventory data, and send the product inventory data to the mobile management terminal to generate a replenishment task instruction;

[0187] The generation module 205 is used to input the RFID tag information and the replenishment task instruction into the replenishment execution unit, generate the replenishment operation data, and update the real-time inventory information;

[0188] The adjustment module 206 is configured to adjust the initial product listing plan based on real-time inventory information and historical sales data to generate a target product listing plan.

[0189] Through the collaborative cooperation of the above components, deep learning and computer vision technologies, in conjunction with the RFID identification system, accurate identification of medical products and real-time inventory monitoring are achieved; the use of a two-stage feature extractor and residual learning method effectively improves the accuracy of product identification in complex environments; by establishing a dynamic product shelving optimization mechanism, intelligent adjustment of product layout and automatic optimization of replenishment strategies are achieved; by introducing mobile management terminals and replenishment execution units, a complete replenishment-acceptance-shelf closed-loop management system is constructed, which significantly improves the operational efficiency and service quality of the medical vending platform; through the intelligent analysis of real-time inventory information and sales data, an adaptive optimization mechanism for the sales model is established, which realizes the rational allocation of product resources and effective control of operating costs.

[0190] above Figure 2The mobile management device for the TCM vending platform in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The mobile management device for the medical vending platform in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0191] Figure 3 1 is a schematic diagram of the structure of a mobile management device for a medical vending platform provided by an embodiment of the present invention. The mobile management device 300 for a medical vending platform may vary significantly due to different configurations or performances, and may include one or more processors (central processing units, CPUs) 310 (e.g., one or more processors) and a memory 320, and one or more storage media 330 (e.g., one or more mass storage devices) storing applications 333 or data 332. The memory 320 and the storage medium 330 may be either short-term storage or persistent storage. The program stored in the storage medium 330 may include one or more modules (not shown), each of which may include a series of instruction operations on the mobile management device 300 for the medical vending platform. Furthermore, the processor 310 may be configured to communicate with the storage medium 330, and execute a series of instruction operations in the storage medium 330 on the mobile management device 300 for the medical vending platform, thereby implementing the steps of the above-mentioned mobile management method for the medical vending platform.

[0192] The medical vending platform mobile management device 300 may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input and output interfaces 360, and / or one or more operating systems 331, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. It will be understood by those skilled in the art that Figure 3 The illustrated structure of the mobile management device for medical vending platforms does not limit the mobile management device for medical vending platforms provided by the present invention and may include more or fewer components than illustrated, or a combination of certain components, or a different arrangement of components.

[0193] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores instructions, which, when executed on a computer, cause the computer to execute the steps of the medical vending platform mobile management method.

[0194] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0195] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc., various media that can store program code.

[0196] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A mobile management method for a medical vending platform, characterized in that: The method comprises: Create an initial product listing plan based on the vending machine operation data of the medical vending platform and collect medical product image data; performing residual learning and two-level feature extraction on the medical product image data to obtain medical product feature data; The medical commodity feature data is input into the medical commodity recognition model to perform medical commodity recognition and obtain the medical commodity recognition result; specifically comprising: inputting the medical commodity feature data into the commodity recognition backbone network, the commodity recognition backbone network adopts the DenseNet structure, and includes four densely connected blocks, each densely connected block is provided with six convolution layers, and dense jump connections are used between adjacent convolution layers to obtain a commodity feature dense map; performing a multi-scale feature fusion operation on the commodity feature dense map, unifying the number of channels of feature maps of different scales to 256 through a 1×1 convolution layer, and using bilinear interpolation to adjust the spatial size of the feature map to the same size to obtain commodity feature fusion data; inputting the commodity feature fusion data into the type recognition branch network, the type recognition branch network includes three parallel attention modules, each attention module is composed of a channel attention unit and a spatial attention unit in series to obtain a commodity type feature vector; inputting the commodity type feature vector into the type classifier, the type classifier includes two A fully connected layer and a Softmax layer, wherein the output dimension of the first fully connected layer is 512, and the output dimension of the second fully connected layer is the same as the number of medical product categories, to obtain product type information; the product feature fusion data is input into the position detection branch network, and the position detection branch network adopts an FPN structure, including a top-down feature pyramid and lateral connections, to obtain a multi-level position feature map; a region proposal calculation is performed on the multi-level position feature map, and three anchor box generators of different scales and non-maximum suppression operations are used to obtain candidate region features, and the dimension of each candidate region feature is 256; the candidate region features are input into the position regressor, and the position regressor includes four layers of fully connected layers, each layer uses a ReLU activation function, and the last layer outputs four coordinate values ​​representing the bounding box coordinates of the product position to obtain product position information; the product type information and the product position information are label aligned, and the product type and position information are associated and matched through the label alignment module to obtain the medical product recognition result; Performing inventory detection based on the medical product identification result to generate product inventory data, and sending the product inventory data to the mobile management terminal to generate a replenishment task instruction; Inputting the RFID tag information and the replenishment task instruction into the replenishment execution unit, generating replenishment operation data, and updating the real-time inventory information; The initial product listing plan is adjusted based on the real-time inventory information and historical sales data to generate a target product listing plan.

2. The mobile management method for medical vending platforms according to claim 1, characterized in that: The process of creating an initial product placement plan based on the vending machine operation data of the medical vending platform and collecting medical product image data includes: Cleaning the sales records of each vending machine in the medical vending platform to obtain valid sales data, and classifying the valid sales data according to product category and sales time to obtain classified sales data; Performing time series analysis on the classified sales data to obtain sales curves for each time period, calculating the sales frequency of commodities in each time period based on the sales curves, and obtaining sales time period parameters; Inputting the sales period parameters into a commodity turnover rate calculation model to obtain commodity turnover speed data, setting safety stock quantities for various commodities based on the commodity turnover speed data, and obtaining replenishment threshold parameters; Performing spatial distribution analysis on the product placement location data of each vending machine to obtain product location popularity data, and prioritizing the product placement locations based on the product location popularity data to obtain location priority data; Performing a correlation analysis on the location priority data and the sales period parameters to obtain location and period matching data, generating a product location distribution matrix based on the matching data to obtain product layout parameters; generating an initial product placement plan based on the replenishment threshold parameter, the product layout parameter, and the sales period parameter, monitoring the remaining quantity of each product in the vending machine based on the initial product placement plan to obtain a product out-of-stock risk value, and determining a priority order for image acquisition based on the product out-of-stock risk value; The merchandise display in the vending machine is spatially mapped according to the merchandise layout parameters to obtain merchandise coordinate data, and the acquisition angle of the image acquisition device is set based on the merchandise coordinate data. The image acquisition device is controlled to perform adaptive angle acquisition according to the image acquisition angle and the image acquisition priority to obtain medical merchandise image data.

3. The mobile management method for medical vending platforms according to claim 2, characterized in that: The performing residual learning and two-level feature extraction on the medical product image data to obtain medical product feature data includes: Performing brightness analysis on the medical product image data to obtain an image brightness distribution matrix, and performing block processing on the medical product image data according to the image brightness distribution matrix to obtain an image block sequence; Input the image block sequence into the backbone feature extraction layer of the residual network, perform feature extraction through three residual units, each residual unit contains two 3×3 convolutional layers and a short-circuit connection, and the convolutional layer uses the ReLU activation function and batch normalization to obtain a backbone feature map; Performing residual mapping calculation on the backbone feature map to obtain a residual feature map, and performing element-level sum operation on the residual feature map and the image block sequence to obtain enhanced image data; Inputting the enhanced image data into the first-level feature extraction network of the two-level feature extractor, wherein the first-level feature extraction network comprises a pyramid structure consisting of five convolutional layers, and a maximum pooling operation is applied after each convolution layer to obtain a multi-scale feature map; Performing channel attention calculation on the multi-scale feature map to obtain a channel weight matrix, and performing weighted processing on the multi-scale feature map according to the channel weight matrix to obtain first-level feature data; Inputting the first-level feature data into the second-level feature extraction network of the dual-level feature extractor, the second-level feature extraction network adopts four parallel dilated convolution branches, and the dilation rates are set to 1, 2, 4, and 8 respectively, to obtain a multi-receptive field feature map; Performing spatial attention calculation on the multi-receptive field feature map to obtain a spatial weight matrix, and performing weighted fusion on the multi-receptive field feature map according to the spatial weight matrix to obtain second-level feature data; The second-level feature data is processed by a feature integration module, which includes two fully connected layers and one Softmax layer to obtain medical product feature data.

4. The mobile management method for medical vending platforms according to claim 3, characterized in that: The performing inventory detection based on the medical commodity identification result, generating commodity inventory data, and sending the commodity inventory data to the mobile management terminal to generate a replenishment task instruction includes: Inputting the product type information and product location information in the medical product identification result into the vending machine inventory detection unit, classifying and statistically processing the product type information to obtain a product category quantity table, and generating a product aisle distribution map based on the product location information; Comparing the commodity category quantity table with the commodity layout parameters in the initial commodity shelving plan to obtain commodity missing data, and establishing a replenishment demand matrix based on the commodity missing data; Performing spatial mapping analysis on the product aisle distribution map and the product layout parameters to obtain product position deviation data, and generating aisle adjustment suggestions based on the product position deviation data; performing priority sorting calculation on the replenishment demand matrix, setting a replenishment level threshold based on the replenishment threshold parameter, and obtaining graded replenishment demand data; Integrate the hierarchical replenishment demand data and the channel adjustment suggestions to obtain commodity inventory data, and encrypt the commodity inventory data using a data encryption module to obtain encrypted inventory data; The encrypted inventory data is sent to the mobile management terminal via a wireless communication module, and identity authentication and data decryption are performed on the mobile management terminal to obtain terminal inventory data; A time-sharing replenishment plan is generated according to the terminal inventory data and the sales period parameters, and replenishment path optimization is performed based on the time-sharing replenishment plan to obtain replenishment path data. The replenishment path data is input into a replenishment scheduling module, and task allocation calculation is performed in combination with personnel configuration information to obtain replenishment task instructions.

5. The mobile management method for medical vending platforms according to claim 4, characterized in that: The step of inputting the RFID tag information and the replenishment task instruction into the replenishment execution unit, generating replenishment operation data, and updating the real-time inventory information includes: Inputting the replenishment path data and the hierarchical replenishment demand data in the replenishment task instruction into a replenishment execution unit, performing task planning for replenishment sites according to the replenishment path data, and obtaining a replenishment execution task sequence; Configuring an RFID reader / writer based on the replenishment execution task sequence, and scanning the RFID tag information of the replenishment product through the RFID reader / writer to obtain the replenishment product tag data; Matching and verifying the replenishment product label data with the product category and quantity table to obtain a product verification result, and generating replenishment execution status data based on the product verification result; Coordinately process the replenishment execution status data and the aisle adjustment suggestions to obtain aisle placement guidance data, and spatially arrange the replenishment products according to the aisle placement guidance data; monitoring the replenishment progress in real time based on the product verification result and the replenishment execution status data, obtaining a replenishment abnormality signal, and comparing the replenishment abnormality signal with the replenishment threshold parameter to generate replenishment warning data; Sending the replenishment warning data to the mobile management terminal via a wireless communication module, performing hierarchical processing on the replenishment warning data, and obtaining a replenishment adjustment plan; Generate replenishment operation data based on the replenishment adjustment plan and the cargo lane layout guidance data, and write the replenishment operation data into a database of the medical vending platform to obtain a replenishment history record; Data fusion processing is performed on the replenishment history records and the commodity layout parameters, the real-time inventory information of the medical vending platform is updated, and the real-time inventory information is fed back to the vending machine inventory detection unit.

6. The mobile management method for medical vending platforms according to claim 5, characterized in that: The adjusting the initial product listing plan based on the real-time inventory information and historical sales data to generate a target product listing plan includes: Associating the real-time inventory information with the replenishment history records according to a commodity category quantity table to obtain commodity circulation data, and performing a time dimension analysis on the commodity circulation data to obtain a commodity sales sample; Constructing a sales location association matrix based on the product sales samples and the product location deviation data, and performing cluster analysis on the sales location association matrix to obtain product regional popularity data; Cross-validating the product area popularity data with the graded replenishment demand data to obtain product turnover rate data, and generating a product zoning configuration plan based on the product turnover rate data; Performing a spatial mapping operation on the commodity zoning configuration plan and the cargo aisle placement guidance data to obtain optimized layout data, and updating commodity layout parameters according to the optimized layout data; Dynamically updating and calculating the replenishment threshold parameter based on the commodity sales rate data to obtain an updated replenishment threshold parameter, and inputting the updated replenishment threshold parameter into a replenishment warning module; Performing a time period association analysis based on the time-sharing replenishment plan and the replenishment execution status data to obtain time period characteristic data, and updating sales time period parameters based on the time period characteristic data; Performing a pattern fusion operation on the updated commodity layout parameters, the updated replenishment threshold parameters, and the updated sales period parameters to obtain commodity sales features; The parameters of the initial product listing plan are updated based on the product sales characteristics to obtain a target product listing plan, and the target product listing plan is written into the operation database of the medical vending platform.

7. A mobile management device for a medical vending platform, characterized in that: The device is configured to execute the medical vending platform mobile management method according to any one of claims 1 to 6, the device comprising: The acquisition module is used to create an initial product listing plan based on the vending machine operation data of the medical vending platform and collect medical product image data; an extraction module, configured to perform residual learning and two-level feature extraction on the medical product image data to obtain medical product feature data; an identification module, configured to input the medical commodity feature data into a medical commodity identification model to perform medical commodity identification and obtain a medical commodity identification result; a detection module, configured to perform inventory detection based on the medical product identification result, generate product inventory data, and send the product inventory data to the mobile management terminal to generate a replenishment task instruction; A generation module, configured to input the RFID tag information and the replenishment task instruction into a replenishment execution unit, generate replenishment operation data, and update real-time inventory information; An adjustment module is used to adjust the initial product listing plan based on the real-time inventory information and historical sales data to generate a target product listing plan.

8. A mobile management device for a medical vending platform, characterized in that: The medical vending platform mobile management device includes: a memory and at least one processor, wherein the memory stores instructions; The at least one processor calls the instructions in the memory to enable the medical vending platform mobile management device to execute the medical vending platform mobile management method according to any one of claims 1 to 6.

9. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the medical vending platform mobile management method according to any one of claims 1 to 6 is implemented.

Images